In 2025, artificial intelligence regulation in the United States is developing faster at the state level than at the federal level. So far, congress has not acted to prevent states from setting their own rules, and as a result, all 50 states have introduced AI-related bills this year.
These bills address topics including government use of AI, healthcare applications, facial recognition, and generative AI.
Some state measures have already taken effect. In Colorado, the Artificial Intelligence Act requires developers to warn about risks when their systems are used for important decisions like hiring or healthcare. In Montana, the Right to Compute law makes AI developers follow safety and security rules for critical infrastructure. And in New York, SB 8755 sets up oversight bodies to watch how AI is used and ensure compliance.
At the federal level, the Trump administration issued its AI Action Plan on July 23, 2025. The plan states that federal AI funding will be withheld from states with regulations deemed overly restrictive, while permitting support for those whose laws do not impose undue constraints.
Congress has not passed comprehensive AI legislation, and an attempt earlier this year to add a clause overriding state rules to a tax and spending bill did not succeed.
Federal AI oversight continues through existing laws, the authority of agencies such as the Federal Trade Commission, and possible adoption of standards created by industry groups through the National Institute of Standards and Technology.
Why It Matters: State and federal governments are currently taking different approaches to AI regulation. States are enacting sector-specific rules focused on transparency, accountability, and risk management, while the federal government is prioritizing AI innovation with fewer new restrictions. The result is a patchwork of legal requirements that vary depending on where an AI system is developed or deployed.
- State Legislative Activity in 2025: Every state introduced AI bills this year, often focusing on disclosure and safe use. Colorado’s AI Act requires transparency for major decisions, Montana’s Right to Compute law sets safety rules for critical infrastructure, and New York’s SB 8755 creates oversight bodies.
- Healthcare and Facial Recognition Laws: In early 2025, 34 states proposed over 250 healthcare AI bills covering uses like insurance decisions and diagnostic tools. By late 2024, 15 states had passed facial recognition laws requiring performance reports and human review of AI results.
- Generative AI Measures: Utah’s Artificial Intelligence Policy Act requires disclosure when generative AI is used for advice or sensitive personal data. Likewise, California’s AB 2013 makes developers share details about the datasets used to train their models. Both laws aim to improve transparency around generative AI.
- Federal Stance and Legislative Efforts: The Trump administration’s AI Action Plan supports AI growth and opposes strict state rules but does not create broad federal oversight. Efforts to override state laws have failed, including a 2023 bill from Senate Majority Leader John Thune that sought reporting rules and certification for high-impact AI.
- Existing legal framework: There is no dedicated federal agency for AI oversight. Regulation comes from the FTC and other sector-specific laws. AI developers do not have Section 230 protections, which could increase legal risk if their systems cause harm.
US regulation of artificial intelligence – Bloomberg
Trusted insights for technology leaders
Our readers are CIOs, CTOs, and senior IT executives who rely on The National CIO Review for smart, curated takes on the trends shaping the enterprise, from GenAI to cybersecurity and beyond.
Subscribe to our 4x a week newsletter to keep up with the insights that matter.
